shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
94,391 Commits 1,712 Branches 298 Tags
Commit Graph

1014 Commits

Author SHA1 Message Date
Jim Laskey
f2516a9180 Added generic code expansion for [signed|unsigned] i32 to [f32|f64] casts in the 
legalizer.  PowerPC now uses this expansion instead of ISel version.

Example:

// signed integer to double conversion
double f1(signed x) {
  return (double)x;
}

// unsigned integer to double conversion
double f2(unsigned x) {
  return (double)x;
}

// signed integer to float conversion
float f3(signed x) {
  return (float)x;
}

// unsigned integer to float conversion
float f4(unsigned x) {
  return (float)x;
}


Byte Code:

internal fastcc double %_Z2f1i(int %x) {
entry:
        %tmp.1 = cast int %x to double          ; <double> [#uses=1]
        ret double %tmp.1
}

internal fastcc double %_Z2f2j(uint %x) {
entry:
        %tmp.1 = cast uint %x to double         ; <double> [#uses=1]
        ret double %tmp.1
}

internal fastcc float %_Z2f3i(int %x) {
entry:
        %tmp.1 = cast int %x to float           ; <float> [#uses=1]
        ret float %tmp.1
}

internal fastcc float %_Z2f4j(uint %x) {
entry:
        %tmp.1 = cast uint %x to float          ; <float> [#uses=1]
        ret float %tmp.1
}

internal fastcc double %_Z2g1i(int %x) {
entry:
        %buffer = alloca [2 x uint]             ; <[2 x uint]*> [#uses=3]
        %tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0                ; <uint*> [#uses=1]
        store uint 1127219200, uint* %tmp.0
        %tmp.2 = cast int %x to uint            ; <uint> [#uses=1]
        %tmp.3 = xor uint %tmp.2, 2147483648            ; <uint> [#uses=1]
        %tmp.5 = getelementptr [2 x uint]* %buffer, int 0, int 1                ; <uint*> [#uses=1]
        store uint %tmp.3, uint* %tmp.5
        %tmp.9 = cast [2 x uint]* %buffer to double*            ; <double*> [#uses=1]
        %tmp.10 = load double* %tmp.9           ; <double> [#uses=1]
        %tmp.13 = load double* cast (long* %signed_bias to double*)             ; <double> [#uses=1]
        %tmp.14 = sub double %tmp.10, %tmp.13           ; <double> [#uses=1]
        ret double %tmp.14
}

internal fastcc double %_Z2g2j(uint %x) {
entry:
        %buffer = alloca [2 x uint]             ; <[2 x uint]*> [#uses=3]
        %tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0                ; <uint*> [#uses=1]
        store uint 1127219200, uint* %tmp.0
        %tmp.1 = getelementptr [2 x uint]* %buffer, int 0, int 1                ; <uint*> [#uses=1]
        store uint %x, uint* %tmp.1
        %tmp.4 = cast [2 x uint]* %buffer to double*            ; <double*> [#uses=1]
        %tmp.5 = load double* %tmp.4            ; <double> [#uses=1]
        %tmp.8 = load double* cast (long* %unsigned_bias to double*)            ; <double> [#uses=1]
        %tmp.9 = sub double %tmp.5, %tmp.8              ; <double> [#uses=1]
        ret double %tmp.9
}

internal fastcc float %_Z2g3i(int %x) {
entry:
        %buffer = alloca [2 x uint]             ; <[2 x uint]*> [#uses=3]
        %tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0                ; <uint*> [#uses=1]
        store uint 1127219200, uint* %tmp.0
        %tmp.2 = cast int %x to uint            ; <uint> [#uses=1]
        %tmp.3 = xor uint %tmp.2, 2147483648            ; <uint> [#uses=1]
        %tmp.5 = getelementptr [2 x uint]* %buffer, int 0, int 1                ; <uint*> [#uses=1]
        store uint %tmp.3, uint* %tmp.5
        %tmp.9 = cast [2 x uint]* %buffer to double*            ; <double*> [#uses=1]
        %tmp.10 = load double* %tmp.9           ; <double> [#uses=1]
        %tmp.13 = load double* cast (long* %signed_bias to double*)             ; <double> [#uses=1]
        %tmp.14 = sub double %tmp.10, %tmp.13           ; <double> [#uses=1]
        %tmp.16 = cast double %tmp.14 to float          ; <float> [#uses=1]
        ret float %tmp.16
}

internal fastcc float %_Z2g4j(uint %x) {
entry:
        %buffer = alloca [2 x uint]             ; <[2 x uint]*> [#uses=3]
        %tmp.0 = getelementptr [2 x uint]* %buffer, int 0, int 0                ; <uint*> [#uses=1]
        store uint 1127219200, uint* %tmp.0
        %tmp.1 = getelementptr [2 x uint]* %buffer, int 0, int 1                ; <uint*> [#uses=1]
        store uint %x, uint* %tmp.1
        %tmp.4 = cast [2 x uint]* %buffer to double*            ; <double*> [#uses=1]
        %tmp.5 = load double* %tmp.4            ; <double> [#uses=1]
        %tmp.8 = load double* cast (long* %unsigned_bias to double*)            ; <double> [#uses=1]
        %tmp.9 = sub double %tmp.5, %tmp.8              ; <double> [#uses=1]
        %tmp.11 = cast double %tmp.9 to float           ; <float> [#uses=1]
        ret float %tmp.11
}


PowerPC Code:

        .machine ppc970


        .const
        .align  2
.CPIl1__Z2f1i_0:                                        ; float 0x4330000080000000
        .long   1501560836      ; float 4.5036e+15
        .text
        .align  2
        .globl  l1__Z2f1i
l1__Z2f1i:
.LBBl1__Z2f1i_0:        ; entry
        xoris r2, r3, 32768
        stw r2, -4(r1)
        lis r2, 17200
        stw r2, -8(r1)
        lfd f0, -8(r1)
        lis r2, ha16(.CPIl1__Z2f1i_0)
        lfs f1, lo16(.CPIl1__Z2f1i_0)(r2)
        fsub f1, f0, f1
        blr


        .const
        .align  2
.CPIl2__Z2f2j_0:                                        ; float 0x4330000000000000
        .long   1501560832      ; float 4.5036e+15
        .text
        .align  2
        .globl  l2__Z2f2j
l2__Z2f2j:
.LBBl2__Z2f2j_0:        ; entry
        stw r3, -4(r1)
        lis r2, 17200
        stw r2, -8(r1)
        lfd f0, -8(r1)
        lis r2, ha16(.CPIl2__Z2f2j_0)
        lfs f1, lo16(.CPIl2__Z2f2j_0)(r2)
        fsub f1, f0, f1
        blr


        .const
        .align  2
.CPIl3__Z2f3i_0:                                        ; float 0x4330000080000000
        .long   1501560836      ; float 4.5036e+15
        .text
        .align  2
        .globl  l3__Z2f3i
l3__Z2f3i:
.LBBl3__Z2f3i_0:        ; entry
        xoris r2, r3, 32768
        stw r2, -4(r1)
        lis r2, 17200
        stw r2, -8(r1)
        lfd f0, -8(r1)
        lis r2, ha16(.CPIl3__Z2f3i_0)
        lfs f1, lo16(.CPIl3__Z2f3i_0)(r2)
        fsub f0, f0, f1
        frsp f1, f0
        blr


        .const
        .align  2
.CPIl4__Z2f4j_0:                                        ; float 0x4330000000000000
        .long   1501560832      ; float 4.5036e+15
        .text
        .align  2
        .globl  l4__Z2f4j
l4__Z2f4j:
.LBBl4__Z2f4j_0:        ; entry
        stw r3, -4(r1)
        lis r2, 17200
        stw r2, -8(r1)
        lfd f0, -8(r1)
        lis r2, ha16(.CPIl4__Z2f4j_0)
        lfs f1, lo16(.CPIl4__Z2f4j_0)(r2)
        fsub f0, f0, f1
        frsp f1, f0
        blr

llvm-svn: 22814
 2005-08-17 00:39:29 +00:00
Chris Lattner
33182325f5 Eliminate the RegSDNode class, which 3 nodes (CopyFromReg/CopyToReg/ImplicitDef) 
used to tack a register number onto the node.

Instead of doing this, make a new node, RegisterSDNode, which is a leaf
containing a register number.  These three operations just become normal
DAG nodes now, instead of requiring special handling.

Note that with this change, it is no longer correct to make illegal
CopyFromReg/CopyToReg nodes.  The legalizer will not touch them, and this
is bad, so don't do it. :)

llvm-svn: 22806
 2005-08-16 21:55:35 +00:00
Nate Begeman
371e49515d Implement BR_CC and BRTWOWAY_CC. This allows the removal of a rather nasty 
fixme from the PowerPC backend.  Emit slightly better code for legalizing
select_cc.

llvm-svn: 22805
 2005-08-16 19:49:35 +00:00
Chris Lattner
1973278b38 Add some methods for dag->dag isel. 
Split RemoveNodeFromCSEMaps out of DeleteNodesIfDead to do it.

llvm-svn: 22801
 2005-08-16 18:17:10 +00:00
Nate Begeman
d5e739dcc2 Fix last night's PPC32 regressions by 
1. Not selecting the false value of a select_cc in the false arm, which
   isn't legal for nested selects.
2. Actually returning the node we created and Legalized in the FP_TO_UINT
   Expander.

llvm-svn: 22789
 2005-08-14 18:38:32 +00:00
Nate Begeman
36853ee1fd Teach the legalizer how to legalize FP_TO_UINT. 
Teach the legalizer to promote FP_TO_UINT to FP_TO_SINT if the wider
  FP_TO_UINT is also illegal.  This allows us on PPC to codegen
  unsigned short foo(float a) { return a; }

as:
_foo:
.LBB_foo_0:     ; entry
        fctiwz f0, f1
        stfd f0, -8(r1)
        lwz r2, -4(r1)
        rlwinm r3, r2, 0, 16, 31
        blr

instead of:
_foo:
.LBB_foo_0:     ; entry
        fctiwz f0, f1
        stfd f0, -8(r1)
        lwz r2, -4(r1)
        lis r3, ha16(.CPI_foo_0)
        lfs f0, lo16(.CPI_foo_0)(r3)
        fcmpu cr0, f1, f0
        blt .LBB_foo_2  ; entry
.LBB_foo_1:     ; entry
        fsubs f0, f1, f0
        fctiwz f0, f0
        stfd f0, -16(r1)
        lwz r2, -12(r1)
        xoris r2, r2, 32768
.LBB_foo_2:     ; entry
        rlwinm r3, r2, 0, 16, 31
        blr

llvm-svn: 22785
 2005-08-14 01:20:53 +00:00
Nate Begeman
180b08897f Some SELECT_CC cleanups: 
1. move assertions for node creation to getNode()
2. legalize the values returned in ExpandOp immediately
3. Move select_cc optimizations from SELECT's getNode() to SELECT_CC's,
   allowing them to be cleaned up significantly.

This paves the way to pick up additional optimizations on SELECT_CC, such
as sum-of-absolute-differences.

llvm-svn: 22757
 2005-08-11 01:12:20 +00:00
Nate Begeman
e5b86d7442 Add new node, SELECT_CC. This node is for targets that don't natively 
implement SELECT.

llvm-svn: 22755
 2005-08-10 20:51:12 +00:00
Chris Lattner
d47675ed24 Eliminate the SetCCSDNode in favor of a CondCodeSDNode class. This pulls the 
CC out of the SetCC operation, making SETCC a standard ternary operation and
CC's a standard DAG leaf.  This will make it possible for other node to use
CC's as operands in the future...

llvm-svn: 22728
 2005-08-09 20:20:18 +00:00
Chris Lattner
96ad31321a Change FindEarliestCallSeqEnd (used by libcall insertion) to use a set to 
avoid revisiting nodes more than once.  This eliminates a source of
potentially exponential behavior.  For a small function in 191.fma3d
(hexah_stress_divergence_), this speeds up isel from taking > 20mins to
taking 0.07s.

llvm-svn: 22680
 2005-08-05 18:10:27 +00:00
Chris Lattner
cabdc34563 Fix a latent bug in the libcall inserter that was exposed by Nate's patch 
yesterday.  This fixes whetstone and a bunch of programs in the External tests.

llvm-svn: 22678
 2005-08-05 16:23:57 +00:00
Nate Begeman
77558da546 Fix a fixme in LegalizeDAG 
llvm-svn: 22661
 2005-08-04 21:43:28 +00:00
Chris Lattner
8191442548 Fix PR611, codegen'ing SREM of FP operands to fmod or fmodf instead of 
the sequence used for integer ops

llvm-svn: 22629
 2005-08-03 20:31:37 +00:00
Chris Lattner
4398daf069 Fix casts from long to sbyte on ppc 
llvm-svn: 22570
 2005-08-01 18:16:37 +00:00
Jeff Cohen
546fd5944e Keep tabs and trailing spaces out. 
llvm-svn: 22565
 2005-07-30 18:33:25 +00:00
Chris Lattner
941d84a34d fix float->long conversions on x86 
llvm-svn: 22563
 2005-07-30 01:40:57 +00:00
Chris Lattner
f59b2daddb Allow targets to have custom expanders for FP_TO_*INT conversions where 
both the src and dest values are legal

llvm-svn: 22555
 2005-07-30 00:04:12 +00:00
Chris Lattner
fe68d75aad Allow targets to define custom expanders for FP_TO_*INT 
llvm-svn: 22548
 2005-07-29 00:33:32 +00:00
Chris Lattner
44fe26ff07 allow a target to request that unknown FP_TO_*INT conversion be promoted to 
a larger integer destination.

llvm-svn: 22547
 2005-07-29 00:11:56 +00:00
Chris Lattner
f99f8f9081 instead of having all conversions be handled by one case value, and then have 
subcases inside, break things out earlier.

llvm-svn: 22546
 2005-07-28 23:31:12 +00:00
Jeff Cohen
5f4ef3c5a8 Eliminate all remaining tabs and trailing spaces. 
llvm-svn: 22523
 2005-07-27 06:12:32 +00:00
Chris Lattner
b35912e421 The assertion was wrong: the code only worked for i64. While we're at it, 
expand the code to work for all integer datatypes.  This should unbreak
alpha.

llvm-svn: 22464
 2005-07-18 04:31:14 +00:00
Nate Begeman
7e74c834c1 Teach the legalizer how to promote SINT_TO_FP to a wider SINT_TO_FP that 
the target natively supports.  This eliminates some special-case code from
the x86 backend and generates better code as well.

For an i8 to f64 conversion, before & after:

_x87 before:
        subl $2, %esp
        movb 6(%esp), %al
        movsbw %al, %ax
        movw %ax, (%esp)
        filds (%esp)
        addl $2, %esp
        ret

_x87 after:
        subl $2, %esp
        movsbw 6(%esp), %ax
        movw %ax, (%esp)
        filds (%esp)
        addl $2, %esp
        ret

_sse before:
        subl $12, %esp
        movb 16(%esp), %al
        movsbl %al, %eax
        cvtsi2sd %eax, %xmm0
        addl $12, %esp
        ret

_sse after:
        subl $12, %esp
        movsbl 16(%esp), %eax
        cvtsi2sd %eax, %xmm0
        addl $12, %esp
        ret

llvm-svn: 22452
 2005-07-16 02:02:34 +00:00
Chris Lattner
e3e847bfd7 Break the code for expanding UINT_TO_FP operations out into its own 
SelectionDAGLegalize::ExpandLegalUINT_TO_FP method.

Add a new method, PromoteLegalUINT_TO_FP, which allows targets to request
that UINT_TO_FP operations be promoted to a larger input type.  This is
useful for targets that have some UINT_TO_FP or SINT_TO_FP operations but
not all of them (like X86).

The same should be done with SINT_TO_FP, but this patch does not do that
yet.

llvm-svn: 22447
 2005-07-16 00:19:57 +00:00
Chris Lattner
f9ddfef872 Fix Alpha/2005-07-12-TwoMallocCalls.ll and PR593. 
It is not safe to call LegalizeOp on something that has already been legalized.
Instead, just force another iteration of legalization.

This could affect all platforms but X86, as this codepath is dynamically
dead on X86 (ISD::MEMSET and friends are legal).

llvm-svn: 22419
 2005-07-13 02:00:04 +00:00
Chris Lattner
ba08a336f0 Fix test/Regression/CodeGen/Generic/2005-07-12-memcpy-i64-length.ll 
llvm-svn: 22417
 2005-07-13 01:42:45 +00:00
Chris Lattner
de0a4b1987 Change *EXTLOAD to use an VTSDNode operand instead of being an MVTSDNode. 
This is the last MVTSDNode.

This allows us to eliminate a bunch of special case code for handling
MVTSDNodes.

llvm-svn: 22367
 2005-07-10 01:55:33 +00:00
Chris Lattner
36db1ed06f Change TRUNCSTORE to use a VTSDNode operand instead of being an MVTSTDNode 
llvm-svn: 22366
 2005-07-10 00:29:18 +00:00
Chris Lattner
0b6ba90a72 Introduce a new VTSDNode class with the ultimate goal of eliminating the 
MVTSDNode class.  This class is used to provide an operand to operators
that require an extra type.  We start by converting FP_ROUND_INREG and
SIGN_EXTEND_INREG over to using it.

llvm-svn: 22364
 2005-07-10 00:07:11 +00:00
Andrew Lenharth
80fe411662 2 fixes: 
1: Legalize operand in UINT_TO_FP expanision

2: SRA x, const i8 was not promoting the constant to shift amount type.
llvm-svn: 22337
 2005-07-05 19:52:39 +00:00
Andrew Lenharth
be3a74ca3e I really didn't think this was necessary. But, Legalize wasn't running again 
and legalizing the extload.  Strange.  Should fix most alpha regressions.

llvm-svn: 22329
 2005-07-02 20:58:53 +00:00
Andrew Lenharth
0a370f4de5 oops 
llvm-svn: 22320
 2005-06-30 19:32:57 +00:00
Andrew Lenharth
b5597e38f6 FP EXTLOAD is not support on all archs, expand to LOAD and FP_EXTEND 
llvm-svn: 22319
 2005-06-30 19:22:37 +00:00
Andrew Lenharth
d74877a46d Adapt the code for handling uint -> fp conversion for the 32 bit case to 
handling it in the 64 bit case.  The two code paths should probably be merged.

llvm-svn: 22302
 2005-06-27 23:28:32 +00:00
Chris Lattner
3268f244e6 allow token chain at start or end of node 
llvm-svn: 22020
 2005-05-14 08:34:53 +00:00
Chris Lattner
865359958b remove special case hacks for readport/readio from the binary operator 
codepath

llvm-svn: 22019
 2005-05-14 07:45:46 +00:00
Chris Lattner
96c262e24b Eliminate special purpose hacks for dynamic_stack_alloc. 
llvm-svn: 22015
 2005-05-14 07:29:57 +00:00
Chris Lattner
669e8c2c9c Use the general mechanism for creating multi-value nodes instead of using 
special case hacks.

llvm-svn: 22014
 2005-05-14 07:25:05 +00:00
Chris Lattner
3eb8693279 legalize target-specific operations 
llvm-svn: 22010
 2005-05-14 06:34:48 +00:00
Chris Lattner
29dcc71d83 LowerOperation takes a dag 
llvm-svn: 22004
 2005-05-14 05:50:48 +00:00
Chris Lattner
d3cc996a47 Allow targets to have a custom int64->fp expander if desired 
llvm-svn: 22001
 2005-05-14 05:33:54 +00:00
Chris Lattner
2e77db6af6 Add an isTailCall flag to LowerCallTo 
llvm-svn: 21958
 2005-05-13 18:50:42 +00:00
Chris Lattner
d0feb64443 Handle TAILCALL node 
llvm-svn: 21957
 2005-05-13 18:43:43 +00:00
Chris Lattner
bb1d60de9c Fix a problem that nate reduced for me. 
llvm-svn: 21923
 2005-05-13 05:17:00 +00:00
Chris Lattner
5a14c8a18e rename variables and functions to match renamed DAG nodes. Bonus feature: 
I can actually remember which one is which now!

llvm-svn: 21922
 2005-05-13 05:09:11 +00:00
Chris Lattner
2a4f7312cd do not call expandop on the same value more than once. This fixes 
X86/2004-02-22-Casts.llx

llvm-svn: 21919
 2005-05-13 04:45:13 +00:00
Chris Lattner
d34cd28aa7 update comment 
llvm-svn: 21916
 2005-05-12 23:24:44 +00:00
Chris Lattner
2dce703710 rename the ADJCALLSTACKDOWN/ADJCALLSTACKUP nodes to be CALLSEQ_START/BEGIN. 
llvm-svn: 21915
 2005-05-12 23:24:06 +00:00
Chris Lattner
111778e665 Pass calling convention to use into lower call to 
llvm-svn: 21900
 2005-05-12 19:56:57 +00:00
Chris Lattner
0bfd177e89 fix expansion of ct[lt]z nodes 
llvm-svn: 21896
 2005-05-12 19:27:51 +00:00